Abstract
An incorporation of spectral voicing information of speech signal for improving the accuracy of automatic phoneme alignment under noisy conditions is presented. Experiments are conducted on the TIMIT speech corpus corrupted by an additive noise at various signal-to-noise ratios. Significant performance improvements are reported.
Full Text 
Sign-in/Register to access full text options
Sign-in/Register to access full text options 
Paper version not known
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
